Cloudera launches blueprints to move AI into production

Cloudera has launched Cloudera Blueprints, a new portfolio that includes NVIDIA AI Blueprints built with NVIDIA Agent Toolkit on Cloudera.

The offering is aimed at companies trying to move artificial intelligence projects from pilot stage into production, particularly where data is spread across cloud environments, on-premises data centres and edge locations.

The portfolio builds on Cloudera's existing Accelerators for ML Projects and adds preconfigured artificial intelligence deployments that run on Cloudera AI. These deployments are designed to connect NVIDIA reference workflows with Cloudera's data management, security and governance tools.

The launch comes as many organisations continue to struggle to turn early artificial intelligence experiments into operational systems. Regulated sectors face particular difficulty because sensitive data often sits across multiple systems and cannot easily be moved.

Industry uses

Initial uses highlighted by Cloudera include transaction foundation models for financial services, healthcare agents for clinical documentation, telecom network configuration planning tools and systems for detecting synthetic video.

In financial services, the technology can help institutions develop models based on transaction data for fraud detection, risk assessment and customer analysis. In healthcare, Cloudera pointed to tools that turn patient-provider conversations into structured clinical notes and support tasks such as intake, scheduling and enquiries.

For telecommunications groups, the software can support radio access network planning by using historical and real-time data to recommend, apply and validate network configurations. Cloudera also identified synthetic video detection as a use case for organisations seeking stronger identity verification and fraud controls.

Leo Brunnick, Chief Product Officer, Cloudera, said: "Enterprises are eager to move beyond AI experimentation, but turning promising prototypes into production-ready applications remains a challenge, particularly when data is spread across complex environments. With Cloudera Blueprints, we're closing that gap by combining proven AI workflows from NVIDIA with the data management, security, and governance capabilities enterprises need to deploy AI at scale. By bringing AI directly to their data, wherever it resides, we're helping organisations accelerate time to value and turn AI innovation into real business impact."

Hybrid deployment

A central part of the launch is support for deployment across public cloud systems, company-run data centres and edge environments. This is intended to let customers develop and run applications without relocating sensitive or regulated data.

Cloudera also outlined the NVIDIA components included in the portfolio. These include NVIDIA NIM microservices and NVIDIA Nemotron models through Cloudera AI Inference, as well as NVIDIA cuDF for Apache Spark in Cloudera Data Engineering.

The combination is intended to support work from data preparation to inference while preserving governance and data lineage across environments. Cloudera said the software is built on an open-source foundation intended to give customers flexibility across models, infrastructure and deployment settings.

The collaboration builds on an existing relationship between Cloudera and NVIDIA as both companies seek a larger role in enterprise artificial intelligence deployments. This latest step focuses on combining NVIDIA's application workflows and computing tools with Cloudera's data management capabilities across hybrid environments.

Pat Lee, Vice President, Strategic Enterprise Partnerships, NVIDIA, said: "NVIDIA and Cloudera share a commitment to helping enterprises accelerate AI adoption and realise the full value of their investments. By combining NVIDIA's accelerated computing and AI software stack with Cloudera's hybrid data platform, we're making it easier for organisations to deploy AI wherever their data resides, whether in the cloud, on premises, or at the edge."

The launch targets a persistent bottleneck in corporate artificial intelligence programmes: the gap between a working prototype and a production system that meets operational, governance and security requirements.

The issue remains significant in markets such as Australia, where the announcement pointed to Deloitte research showing only 28% of enterprises have moved at least 40% of their artificial intelligence pilots into production.